Estimating the center of mass position and the angular momentum derivative of legged systems is essential for both controlling legged robots and analyzing human motion. In this paper, we introduce a new recursive approach to accurately estimate these two quantities together, by fusing the kinetic and kinematic measurements coming from classic sensors available in robotics and biomechanics, and by exploiting the accuracy of each measurement in the spectral domain. The soundness of the proposed approach is first validated on a simulated humanoid robot, where access to ground truth data is granted. The results show that our method reduces the estimation error on the center of mass position with regard to kinematic estimation alone, while also providing an accurate estimation of the derivative of angular momentum. We finally illustrate the effectiveness of our method on real measurements coming from walking experiments on the HRP-2 humanoid robot.